High variation in the signal displayed by MNE-Qt-Browser as one moves through the signal

💬 Support & Discussions
1

Hi Folks,

I’m writing to hear your thoughts about something we’ve noticed while using the Qt browser. The displayed signals change substantially when one scrolls just a few seconds through the signal. Here’s a brief GIF to show what I’m talking about.

Is this normal behavior? Is there any way the variation can be reduced?

mne-qt-browser-behavior

Many thanks!

2

I don’t really see an issue here.
You could crop the end of the recording where it looks like your EEG cap was unplugged from the amplifier creating a large discontinuity. I’m guessing this large discontinuity is affecting the filter you applied to your data.

If you do:

from matplotlib import pyplot as plt

plt.plot(raw.times, raw.get_data(picks=0).squeeze())

You should see the same plot (for channel 0), (except if you applied visualisation filters with raw.plot(lowpass=..., highpass=...)).

EDIT: @ealtamir I now see your issue. Please share the version of the qt-browser and of mne used (use mne.sys_info()); I believe this issue was fixed already and you are running on outdated versions.

Mathieu

2 Likes
3

Hi Mathieu,

Thanks for your reply. I’m using v0.6.0, I will try the latest version as per your recommendation. For reference, he’s the output of mne.sys_info():

Platform             Windows-10-10.0.22631-SP0
Python               3.11.2 (tags/v3.11.2:878ead1, Feb  7 2023, 16:38:35) [MSC v.1934 64 bit (AMD64)]
Executable           C:\Users\phonb\AppData\Local\pypoetry\Cache\virtualenvs\pipeline-gui-3-Rle7Yj-py3.11\Scripts\python.exe
CPU                  Intel64 Family 6 Model 151 Stepping 2, GenuineIntel (24 cores)
Memory               31.7 GB
Core
├ mne               1.6.0 (outdated, release 1.6.1 is available!)
├ numpy             1.26.3 (OpenBLAS 0.3.23.dev with 24 threads)
├ scipy             1.11.4
├ matplotlib        3.8.2 (backend=QtAgg)
├ pooch             1.8.0
└ jinja2            3.1.2
Numerical (optional)
├ sklearn           1.3.2
└☐ unavailable       numba, nibabel, nilearn, dipy, openmeeg, cupy, pandas
Visualization (optional)
├ qtpy              2.4.1 (PySide6=6.4.2)
├ pyqtgraph         0.13.3
├ mne-qt-browser    0.6.0
└☐ unavailable       pyvista, pyvistaqt, vtk, ipympl, ipywidgets, trame_client, trame_server, trame_vtk, trame_vuetify
4

This version should be recent enough… we might have a bug on our hands. Could you also try:

raw.plot(precompute=False)
# vs
raw.plot(precompute=True)

Mathieu

5

I don’t understand what the problem is from the movie, can you explain?

6

@cbrnr Look at the left side (white background), after an horizontal scroll, the traces change completely. It kind of reminded me of a bug with projectors being toggle on/off during horizontal scrolls.